As recognized herein, scrolling of content presented on a display may be performed by a device based on input from a camera. For example, eye movements and head tilts as detected using the camera may be used for the device to determine that it should scroll text as a user reads through the text so that the user does not need to provide touch input to the display to scroll but is still able to continue reading the text. This may be desirable if the user is, for instance, reading instructions for cooking or working on an automobile and the user's hands are dirty, and hence the user might not wish to touch the device with dirty hands. However, as also recognized herein, there are instances where this type of scrolling may also be annoying, such as when the user is holding the device in his/her hand and simply wishes to scroll manually by directing input to the device's display. There are currently no adequate solutions to the foregoing computer-related, technological problem.